Paper by E Acosta, @demography.bsky.social, @thegargiulian.bsky.social & C Torres first from PDR special issue I guest edited w/ L Andriano & M Ebbinghaus, Social and Demographic Consequences of Political Conflict and Violence, to be available on early view onlinelibrary.wiley.com/doi/10.1111/...
Weaponizing Kinship: Four out of ten Colombians have lost a family member in the conflict 🇨🇴
Enrique Acosta, Diego Alburez, Maria Gargiulo, & Catalina Torres quantified how many Colombians have ever lost at least a family member to conflict-related violence.
www.demogr.mpg.de/go/weaponkinship
What happens when the patterns we’ve studied in other countries start showing up closer to home?
HRDAG’s 2025 review reflects on how the rise of tyranny in the United States is changing our work and our threat model.
Read it here: hrdag.org/2026/03/03/h...
blog post: sampling to assess data quality
@bhedtgauthier.bsky.social et al. (2012) used sampling to assess and improve data quality in Malawi
Congratulations to Ayyub Ibrahim, recipient of the Electronic Access award from the Society of Professional Journalists, Northern California! spjnorcal.org/2026/02/14/s...
👉 Read why bias in data is nearly everywhere (and what it means for truth in data): hrdag.substack.com/p/yes-your-d...
White text on orange: "The crucial bit of magic that makes the entire field of statistics work is understanding what the sample statistics is telling you about the population parameter."
Understanding the relationship between the data we have and the world we are researching is at the heart of statistics.
hrdag.substack.com/p/yes-your-d...
Among the many ways HRDAG contributed in post-conflict Guatemala: we trained Guatemalan NGOs in methods for securing their data using open source information management tools.
hrdag.org/guatemala/
HRDAG and partners gather around a large conference table with laptops for a data retreat.
HRDAG hosted a data retreat in our office last week.
We're grateful to be part of a community of statisticians and scientists using data to advance human rights.
🎉 ¡Abrió el nuevo llamado al Programa de Campeones de rOpenSci!
Durante las próximas semanas vamos a compartir una serie de videos con testimonios de participantes de ediciones anteriores.
En primera persona, cuentan cómo fue su experiencia y qué impacto tuvo el programa en sus proyectos y […]
HRDAG statisticians and friends gathered around a table for a data retreat with laptops.
HRDAG statisticians and friends gathered around a table for a data retreat with laptops.
HRDAG statisticians and friends gathered around a table for a data retreat with laptops.
This work is better with friends.
Here are some moments from our recent data retreat with HRDAG scientists and our partners at @datacivica.bsky.social Kilómetro 0, & @invisible.institute
PSG seminars continue 📈
Up next on Jan 28 at 12:45, Research Fellow @jacalazans.bsky.social will give a talk on how tobacco control policies influenced adult mortality patterns in Brazil.
Join us in person (LG24, Keppel Street) or online. Find more details 👇
www.lshtm.ac.uk/newsevents/e...
More data doesn’t automatically mean more knowledge.
That's because raw data is inherently biased and often misrepresents reality.
Statistical models help reveal those biases and uncover the truth.
www.youtube.com/watch?v=z7U9...
blog post: Total Margin of Error
margin of error = 2 x (reported margin of error)
and how much of this error is "bias" vs "variance" ?
The ICC is a powerful tool against impunity. That's why HRDAG adds our voice to the community of human rights organizations demanding that the Trump Administration respect international law and the ICC’s essential independence.
hrdag.org/2026/01/05/t...
blog posts: should we adjust for a mismeasured X ?
You know the population distribution for X (e.g. vote choice in 2024).
But you only have a reported X* in your survey.
Should you adjust for it ?
Later today: exploring toy examples to see.
From the archives: learn about the statistical concepts of "capture-recapture," also known as "multiple systems estimation."
hrdag.org/tech-notes/b...
White text on multicolored background: Accompaniment is a practice taht emerged in response to violence against human rights defenders in conflict zones.
Accompaniment - acting as a witness and support to human rights advocates in conflict zones - is core to HRDAG's history and origin story.
Today, those principles still guide our work in data and statistical reasoning. hrdag.substack.com/p/all-of-the...
blog post: 3rd helpings of the logit shift
You have multiple outcomes, but only some have aggregate truth to shift to.
How can we calibrate our estimates of p(y_1, y_2 | X) to aggregate data about E[y_1] ?
@wpmarble.bsky.social and Josh Clinton have ideas !
HRDAG is part of a global community of human rights partners, and collaboration is key to everything we do. Today we celebrate one of our long-time partners, Valentina Rozo Angel, joining the Global Rights Innovation Lab Clinic at @ucberkeleylaw.bsky.social www.law.berkeley.edu/article/expa...
White text on multicolored card including the words: RightsX Summit: The future of human rights starts here. Solution Roundtable: Pushing the limits | Strengthening synergies between Citizen Data, Ai & Official Statistics.
Join HRDAG's Maria Gargiulo at RightsX Summit to explore how citizens and human rights groups use data to seek justice. Registration here: rightsx.ohchr.org
The HRDAG staff participating in a candle light vigil.
Today and every day, we offer gratitude to our donors and funders. Thanks to your support, we use science to defend truth and human rights.
Today we express gratitude to all the survivors who shared their stories with our research teams, from Guatemala to South Africa to the United States and beyond, as we researched and analyzed state violence.
Your stories help pave the way for justice and accountability.
blog post: quantity vs quality
compare 2 surveys:
1. 100% coverage, but response probability P[R = 1 | Y] differs a lot by Y
2. Only 5% coverage, but P[R = 1 | Y] is roughly constant across Y
which would you use ? both ?
“As researchers and research activists, it’s our responsibility to question what gets measured and reported and how” - @thegargiulian.bsky.social discusses the importance of mortality estimation research
So excited to see this piece out in Demography. We find that exposure to community-level homicides is associated with higher probabilities of contraceptive first use and provision.
This paper forms the basis of my ERC Starting Grant on how local violent crime affects SRH in Mexico. 🇲🇽 More to come!
Want better AI? Involve human experts.
New research from HRDAG and Cornell's CAT Lab offers powerful evidence for how participatory models improve reliability and validity. hrdag.org/2025/11/17/n...
new blog post: sampling the sample
we’ve focused on estimating means E[Y].
but say Y are openends ("describe how you feel about the candidate") and you want to read thru a few draws from the population, not only survey responders.
what should you do ?
HRDAG's executive director Megan Price teamed up with CAT Lab Founder @natematias.bsky.social to research how human participation can improve AI. Read our latest findings. hrdag.org/2025/11/17/n...
New article in @pnas.org: when scientists bring in the public, it can directly improve the science of AI. hrdag.org/2025/11/17/n...
